(i) Field of the Invention
The present invention relates to a method for identifying putative peptides from nucleotide or peptide sequences of unknown function such as both nucleic acid and peptide precursors of a peptide comprising an amidated C-terminal end and, more particularly, to a method wherein putative precursor peptides are identified from a genetic database.
(ii) Description of Related Art
Certain combinations of nucleotides, when present in a polynucleotide, are known to give rise to certain properties in the polypeptide translated therefrom. One example includes those nucleotides which encode polypeptide hormone precursors which undergo a post-translation amidation reaction. Another example, as set forth in U.S. Pat. No. 4,917,999, relates to certain nucleotides which are characteristic of polypeptides exhibiting xcex1-amylase enzymatic activity.
Amidated polypeptide hormones are synthesized in the form of a precursor which undergoes maturation. This maturation consists of an amidation reaction. The amidation reaction of the C-terminal end is a characteristic reaction of amidated polypeptide hormones. This reaction, which occurs on the precursor of one or more hormones, allows maturation of the hormone and also ensures its biostability in the physiological medium: the amide group formed is less vulnerable than the free acid function. The hormone is therefore more resistant to carboxypeptidases, it remains active in the cell for longer and retains an optimum affinity for its receptor site.
Amidation has been widely described (xe2x80x9cPeptide amidationxe2x80x9d, Alan F. Bradbury and DerekG. Smyth, TIBS 16:112-115, March 1991 and xe2x80x9cFunctional and structural characterization of peptidylamidoglycolate lyase, the enzyme catalysing the second step in peptide amidationxe2x80x9d, A. G. Katopodis, D. S. Ping, C. E. Smith and S. W. May, Biochemistry, 30(25): 6189-6194, June 1991), and its mechanism is as follows:
1xe2x80x94Cleavage of the precursor polypeptide chain of the hormone by an endoprotease at the two basic amino acids, that is to say arginine and/or lysine,
2xe2x80x94Subsequently two cleavages by carboxypeptidase result, which lead to the extended glycine intermediate,
3xe2x80x94The enzyme PAM (peptidyl-glycine-amidating monooxygenase) comprises two distinct enzymatic activities: firstly, it converts the extended glycine intermediate into an xcex1-hydroxyglycine derivative, the subunit of the enzyme PAM involved is PHM (peptidyl-glycine-hydrolylating monooxygenase). The derivative obtained serves as the substrate for the second subunit of PAM (called PAL: peptidyl-hydroxyglycine-amidating lyase), which fixes the amine function of the glycine on to the amino acid immediately adjacent to the N-terminal side and liberates glyoxylate.
This reaction involves the presence of a recognition site on the precursor of the hormone or hormones, a site which always comprises the sequence: glycine and two basic amino acids (arginine or lysine). The amidated polypeptide hormones which are to be secreted outside the endoplasmic reticulum are known to comprise a consensus signal sequence of about fifteen to thirty amino acids, this sequence being present at the N-terminal end of the polypeptide chain. It is cut later by a signal peptidase enzyme such that it is no longer found in the protein once secreted.
Given the importance of known amidated polypeptides in the context of numerous biological systems, methods have been sought for the identification of additional amidated polypeptides. Unfortunately, at the present time, the discovery of a new protein is not easy.
To date, the art has developed certain approaches in an attempt to identify novel proteins of potential biological interest.
In one approach, potentially new proteins of interest are isolated from a source by selecting a specific property which the researcher believes will be possessed by one or more potential proteins of interest in a sample. According to this approach, proteins can be isolated and purified by various techniques: precipitation at the isoelectric point, selective extraction by certain solvents and then purification by crystallization, counter-current distribution, adsorption, partition or ion exchange chromatography, electrophoresis.
The conventional protein isolation techniques described above provide only limited success in the isolation and identification of new biological molecules of interest. This approach implies knowledge of the properties of the protein to be isolated. Typically, one of two situations arises based on isolation of proteins using a common property. In the first situation, the common property will be for the most part unrelated or only marginally related to the biological function of the molecules being isolated. One could envision, for example, two proteins sharing identical isoelectric points but having completely unrelated biological functions. In the second situation, separation might be achieved based on common property which is very closely related to the biological function of the molecule being isolated. In this category, for example, one might envision molecules which bind to the same receptor molecule. In the former situation, the isolation of potentially new polypeptides is quite unfocussed given its likelihood of isolating compounds of completely unrelated biological function. By complete contrast, the latter situation suffers the exact opposite deficiency in that it enables isolation of only a very limited number of new biologically interesting molecules.
Thus, a person skilled in the art seeking to isolate potentially new polypeptides of interest by conventional protein separation techniques was confronted with the dilemma of obtaining a hodgepodge of biologically unrelated polypeptides or, alternatively, only a very specific set of polypeptides.
Another serious shortcoming of conventional techniques for isolation of new polypeptides from a sample relates to the nature of the sample itself. Obviously, there will be a limited number of available polypeptides for isolation and identification in any given biological sample. Furthermore, great care must be taken with such samples to ensure the continued integrity of the biologically active molecules therein.
Not surprisingly, previous attempts to isolate and characterize new peptides comprising an amidated C-terminal end have followed the conventional approach of starting with a biological sample and choosing from the arsenal of known separation techniques for isolating and identifying the peptides. For example, in U.S. Pat. No. 5,360,727 in the name of Matsuo et al., there was isolated a C-terminal alpha-amidating enzyme of porcine origin by extracting and purifying the enzyme from porcine atrium cords exhibiting the enzyme activity. In U.S. Pat. No. 5,871,995 issued in the name of lida et al., purified enzymes participating in C-terminal amidation were purified from a biological material such as horse serum by affinity chromatography using a peptide C-terminal glycine adduct as a ligand. In U.S. Pat. No. 4,708,934 in the name of Gilligan et al., peptidyl-glycine alpha.-amidating monooxygenase enzyme was extracted from medullary thyroid carcinoma cell lines and tissue samples. Where identification of substantial numbers of new polypeptides capable of amidation is the goal, conventional isolation techniques such as these are completely unsuitable, as they typically permit isolation of only a single polypeptide of interest from a source suspected to contain that polypeptide.
In PCT/FR98/01767, the assignee of the present application has recently developed a method which overcomes many of disadvantages discussed above in that it enables the rapid identification of a large number of putative peptides which comprise an amidated C-terminal end. In particular, unlike earlier techniques which relied on a particular physical property of the polypeptide to isolate it from a source suspected or known to contain it, the method developed by the assignee relies on a characteristic of the peptide sequence of the precursor of all amidated hormones known to date, thereby allowing simultaneous detection of several new hormones of this category. More particularly, this technique relies on the direct identification of the nucleotide sequence which codes for the precursors in cDNA banks prepared from tissues in which the precursors of these hormones can be synthesized.
The method of PCT/FR98/01767 permits identification of the precursor of a peptide having an amidated C-terminal end, by the following successive stages:
1xe2x80x94Obtaining of a DNA bank;
2xe2x80x94Hybridization of one or more oligonucleotides OX with the DNA bank;
3xe2x80x94Identification of the DNA sequence or sequences of the bank which hybridizes with an oligonucleotide OX;
4xe2x80x94Identification in this sequence or sequences of one or more peptides with a possible amidated C-terminal end.
OX is a single-stranded oligonucleotide which can hybridize under mild conditions with an oligonucleotide OY of the sequence Y1-Y2-Y3-Y4-Y5, in which Y1 represents a nucleotide sequence of 1 to 12 nucleotides or Y1 is suppressed, Y2 represents a trinucleotide which codes for Gly, Y3 and Y4 independently represent a trinucleotide which codes for Arg or Lys and Y5 represents a nucleotide sequence of 1 to 21 nucleotides or Y5 is suppressed.
Preferably, the DNA bank is a cDNA bank. A cDNA bank contains the cDNA corresponding to the cytoplasmic mRNA extracted from a given cell. The bank is called complete if it comprises at least one bacterial clone for each starting mRNA. Hybridization takes place if two oligonucleotides have substantially complementary nucleotide sequences, and they can combine over their length by establishing hydrogen bonds between complementary bases.
The search by this method of PCT/FR98/01767 has been found to be much less restricting than the abovementioned conventional techniques of biochemistry, since:
it can lead to the isolation of several distinct precursors present in the same tissue by the same principle;
it allows detection, under the same technical conditions, of precursors corresponding to hormones which have very different biochemical and biological properties;
it allows concomitant identification of all the peptide hormones which can be contained in the same precursor.
As a result, the screening technique set forth in PCT/FR98/01767 allows a not insignificant saving in time and money in a sector where the costs of research and development represent a very high proportion of turnover.
By allowing the obtaining of a large number of potentially therapeutically useful polypeptides, the technique developed allows pharmacological study of active substances having a fundamental physiological roll in the mammalian organism: hormones and more particularly amidated polypeptide neurohormones. Having available for the first time cDNA corresponding to active substances, it is now possible to introduce the cloned vector by genetic engineering to lead to synthesis of hormones having a therapeutic use by means of microorganisms.
Although giving rise to numerous significant advantages in terms of the ability to rapidly obtain a large number of putative candidate peptide molecules which serve as precursors to peptides comprising an amidated C-terminal end, there are nonetheless still certain difficulties attendant with the use of a cDNA bank for carrying out the screening. As discussed earlier, the cDNA bank typically derives from a single cell and therefore will contain only those polypeptides which are expressed in that cell. This means that even if within the genome of the cell, the screening method will not detect a putative peptide if that peptide is not expressed in that cell. Furthermore, even to the extent that a polypeptide of interest is expressed in the cell, the screening technique is necessarily limited to polypeptides expressed by that particular cell and, indeed, by the particular species of life from which the cell is derived. This makes it difficult to screen for the vast numbers of putative peptides which are of interest.
Thus, the method of the PCT/FR98/01767 solved a very important restriction in the identification of putative peptides serving as precursors of peptides comprising an amidated C-terminal end. More specifically, while the method of PCT/FR98/01767 certainly provides the means through which to probe a cDNA bank for all possible sequences having the desired post-translational amidation property, it nonetheless was restricted to those cDNAs found in the cDNA bank.
Recently, there has been an interest in using available databases containing vast numbers of nucleotide sequences in order to, for example, compare a sequence of interest with known sequences.
For example, in U.S. Pat. No. 5,706,498, there is disclosed a gene database retrieval system for making a retrieval for a gene sequence having a sequence similar to a sequence data from a gene database. The gene database stores the sequence data of genes whose structures or sequences have already been analyzed and identified. The system includes a dynamic programming operation unit for determining the degree of similarity between target data and key data by utilizing the sequence data of the bases of the gene from the gene database as the target data and the sequence data of the bases as the key for retrieval, and a central processing device unit for executing the access process to make access to the gene database, in parallel to the operation process for determining the degree of similarity by transmitting the sequence data of the bases from the gene database continually one after another into the dynamic programming operation unit as the target data, by controlling the gene database and the dynamic programming operation unit.
U.S. Pat. No. 5,873,082 discloses a sequence database search wherein a homologous sequence of a given sequence is searched from the sequence database and the results outputted in order of higher homology. According to the patent, a plurality of lists having similarities and differences can be effectively compared. In the case of the sequence database search results, a large number of lists including a huge number of sequence names can be quickly compared.
In U.S. Pat. No. 5,577,249, there is disclosed a method for finding a reference sequence in a database. The most preferred embodiment has specific application to searching the genome of living organisms, in particular the human genome, to find locations and purposes of nucleotide sequences and other biological information that are found on strings of DNA. The method employs human genome databases commercially available which have substrings of the DNA chains broken down into nucleotide token sequences. A unique original index associated with the original DNA string is then created. A reference nucleotide sequence is selected. The reference indexes and original indexes are compared. The method was applied to match reference strings of nucleotides for the genome of E. coli which contains approximately 4 million nucleotides.
U.S. Pat. No. 5,701,256 discloses a method and apparatus for sequence comparisons wherein new proteins sequences are compared with known sequences, such as from a sequence database, typically with a view to determine what level of similarity is shared between the proteins in terms of structural and functional characteristics.
U.S. Pat. No. 5,523,208 discloses a method for scanning nucleotide or DNA sequence date banks to identify genetic regions or genes coding for biologically interacting proteins. In particular, the method provides a means of scanning data banks consisting of cloned genetic material, including but not limited to DNA, RNA, mRNA, tRNA and nucleotide fragments, to identify the function of genetic material of unknown function. The method, when used on DNA fragments of unknown coding potential will produce a list of gene fragments which code for proteins having the potential to form complexes or multimeric configurations with the unknown protein.
As the above discussion demonstrates, one of the major applications for computerized methods of searching databases of genetic information is the comparison of a newly found sequence of unknown function with a database of sequences of known function. The goal of such methods of course is to ascertain the function of the unknown protein by relating it to structurally similar proteins of known function. Another application of computerized methods is to find those sequences in a database which are structurally related. Still further methods try to find sequences which form complexes with a known sequence. The focus of all of these methods is generally on comparison of one sequence with another sequence with a view to determining which sequences are structurally similar, rather than determining which genes, from among a large database of genes, possesses a given biological property even where such genes are generally not structurally similar.
The present invention has as its primary objective the overcoming of the disadvantages attendant with prior art techniques for identifying biologically interesting molecules from among a large group of candidate molecules. This objective is achieved by providing a method for identifying putative peptides of a given function from among nucleotide or peptide sequences of unknown function by screening a database for the presence of a particular combination of nucleotides or amino acids indicative of the peptides of given function.
It is a further object of the present invention to provide a method of identifying a putative peptide of a given function which is not limited to those peptides expressed in a particular biological source, for example because the protein is not expressed in that source.
It is another object of the present invention to provide a method of identifying a putative peptide of a given function which does not depend on physical properties of the peptide, such as isoelectric point or solubility, for the identification.
Yet another object of the present invention is to provide a method of identifying a putative peptide of a given function which is applicable to proteins which otherwise are biologically unrelated in their physical properties.
Still another object of the present invention is to provide a method of identifying a putative peptide of a given function from among candidate polypeptides which exhibit a very low degree of homology with each other.
Another object of the present invention is to provide a method of identifying a putative peptide of a given function which facilitates pharmacological study of active substances having a fundamental physiological role in an organism such as hormones and, more particularly, amidated polypeptide hormones.
Another object of the present invention is to provide a method of identifying a putative peptide of a given function which can be carried out with available genetic databases and available software.
Briefly described, these and other objects of the invention are achieved by providing a method for identifying putative peptides of a given function from among nucleotide or peptide sequences of unknown function comprising the steps of:
(i) obtaining a polynucleotide or polypeptide database;
(ii) screening said database for the presence of a combination of nucleotides or amino acids indicative of the peptide of given function;
(ii) identifying the polynucleotide or polypeptide sequences which comprise the combination of nucleotides or amino acids indicative of the peptide of given function.
In a preferred aspect, the present invention provides a method for identifying a precursor of a peptide comprising an amidated C-terminal end comprising the steps of:
(i) obtaining a polynucleotide or polypeptide database;
(ii) screening the database for the presence of a combination of nucleotides or amino acids indicative of the precursor of the peptide comprising the amidated C-terminal end;
(ii) identifying the polynucleotide or polypeptide sequences which comprise the combination of nucleotides or amino acids indicative of the precursor of the peptide comprising the amidated C-terminal end.
The database preferably comprises polynucleotide sequences and/or polypeptide sequences corresponding to the polynucleotide sequences and, optionally, accession numbers for the polynucleotide sequences. Where the polypeptide sequences are not available in the data base, they may be obtained by translating the polynucleotide sequences in said database. In a preferred embodiment, three different polypeptide sequences are obtained, corresponding to translation of three different reading frames of said polynucleotide sequences.
The database may further include annotational information relating to the polypeptide or polynucleotide sequences, such as at least one of origin, source, features and references for the sequences.
In screening the database, it is preferred to first locate the AUG start codon in a polynucleotide sequence and verify that no stop codon is present between the AUG start codon and the combination of nucleotides indicative of the precursor of the peptide comprising the amidated C-terminal end.
Once the step of identifying nucleotide or polypeptide sequences has been carried out, the identified sequences can be compared with sequences of known biological function and those identified sequences whose biological function is unknown selected. Alternatively, after the step of identifying nucleotide or polypeptide sequences has been carried out, the similarity of the selected sequences of unknown biological activity can be compared with sequences of known function and, if no similar sequence is found, the sequence of unknown biological activity selected for further investigation. If a similar sequence is found, the sequence of unknown biological activity can be selected as a candidate sequence exhibiting the putative function of the known similar sequence.
In one embodiment, the identified polypeptide sequence can be obtained and the properties of the polypeptide sequence evaluated.
The combination of nucleotides preferably comprises the sequence Y1-Y2-Y3-Y4-Y5, in which Y1 is a nucleotide sequence of 1 to 12 nucleotides or is suppressed, Y2 is a codon for Gly, Y3 and Y4 independently are codons for Arg or Lys and Y5 is a nucleotide sequence of 1 to 21 nucleotides or Y5 is suppressed.
With the foregoing as well as other objects, features and advantages of the invention that will become hereinafter apparent, the nature of the invention may be better understood by reference to the Detailed Description of the Preferred Embodiments and to the appended claims.
By xe2x80x9cputative peptides of a given functionxe2x80x9d is meant polypeptides which include a particular oligonucleotide sequence which is characteristic of a particular function shared by many proteins among a single species and/or among different species. In particular, certain oligonucleotide sequences have been found to be associated with certain types of proteins, such as C-terminal amindated hormones or amylases. The invention is applicable wherever there is an oligonucleotide sequence indicative of such a function.
By xe2x80x9cprecursor of a peptide comprising an amidated C-terminal endxe2x80x9d is meant any of the precursor proteins which undergo an amidation reaction at the C-terminal end as described, for example, in Bradbury et al.
By xe2x80x9cpolynucleotide or polypeptide databasexe2x80x9d is meant any of the publicly available databases, such as FASTA, GENBANK, PROSITE or SWISS-PROT which typically include polynucleotide and/or polypeptide sequence data. The nucleotide sequence data is available, for example, at EMBL, GENBANK and at other places such as EXPASY. Such data often will also include ACCESSION numbers.
If the peptide sequences corresponding to a given ACCESSION number are available (e.g., in SWISS-PROT), then this xe2x80x9cvalidatedxe2x80x9d sequence is included in the database. Otherwise, the nucleotide sequence is preferably translated using programs available in the art such as Translate and/or Back-Translate. The translation is carried out using the data associated with the nucleotide sequence (3xe2x80x2 or 5xe2x80x2 end) and three different reading frames (N, N+1, N+2). If this information is not available, the translation is carried out using both the available nucleotide sequence and its complementary sequence (six putative peptides for a single nucleotide sequence).
Optionally, such database further includes annotations containing all of the information that is available for a particular sequence, such as ORIGIN, SOURCE, FEATURES, related REFERENCES and COMMENTS associated with the sequence. This facilitates further database mining. An example of the available for annotation is given below:
By xe2x80x9ccombination of nucleotides or amino acids indicative of the precursor of the peptide comprising the amidated C-terminal endxe2x80x9d is preferably meant the sequence Y1-Y2-Y3-Y4-Y5, in which Y1 is a nucleotide sequence of 1 to 12 nucleotides or is suppressed, Y2 is a codon for Gly, Y3 and Y4 independently are codons for Arg or Lys and Y5 is a nucleotide sequence of 1 to 21 nucleotides or Y5 is suppressed. The particular combination is set forth in detail in PCT/FR98/01767, the disclosure of which is hereby incorporated by reference.
In the preferred embodiment, the database to be screened is obtained by running the sql/plus script follows:
Obviously, the size of the different ORACLE fields can be adjusted in order to fit the size of the data that is being imported in the tables.
The SOLOCUS, SODEFINITION, SOACCESSSION, SOORGANISM, SONID, SOKEYWORDS, SOSOURCE, SOREFERENCE, SOCOMMENT, SOFEATURES, SOBASECOUNT, SOSEQUENCE fields in oracle contains respectively the LOCUS, DEFINITION, ACCESSSION, ORGANISM, NID, KEYWORDS, SOURCE, REFERENCE, COMMENT, FEATURES, BASECOUNT , SEQUENCE entries of the genbank data files.
The SOPHASE contains the reading frame (0, +1, +2) if the peptide sequence is generated using IHB""s DNA translator, SOPEPTIDE contains the peptide sequence corresponding to a given SOACCESSION, finally SOPEPTIDEORIGIN tells how the peptide sequence have been obtained (IHB""s DNA Translator, GENBANK, SWISS-PROT, . . . ).
When a FASTA formatted file (see below) is used instead of a GENBANK formatted file, the header line is stored in the SODEFINITION field when we are dealing with a sequence of nucleotides and the nucleotidic sequence itself is stored in the SOSEQUENCE field. If the data being inserted in the database is a peptide sequence then the header line is stored in the SOPEPDEFINITION and the peptidic sequence is stored in the SOPEPTIDE field once again the SOPEPTIDEORIGIN will tell where the peptide sequence comes from (GENBANK, SWISS-PROT, . . . ).
.gi|402336 (M17352) dnaN protein [Salmonella typhimurium]
MKFTVEREHLLKPLQQVSGPLGGRPTLPILGNLLLQVADGALSLTGTDLEMEMVARVTLSQP SEQ ID NO:8
 greater than gi|306148 (Li19604) core polypeptide [Heliobacillus mobilis]
MATADAAFNPRAQVFEWFKDKVPATRGAVLKAHINHLGMVAGFVSFVLVHHLSWLSDQVLFAPTPIFYARLYQLGLDASARSADALMVARLHLPAAIIFWIIGHIKTPREDEFLKNVTFGKTLVAQFHFLALVATLWGMHMAYIGVRGANGGIVPTGLSFDMFGPITGATLAGNHVAFGALLFLGGVFHHFAGFNTKRFAFFEKDWEAVLSVSAQVLAFHFATVVFAMIIWNRPDQPILSFYFMQDYALSNYAAPEIREIASQNPGFLIKQVILGHLVFGVMFWIGGVFHGASLHVRATNDPKLAEALKDFKMLKRCYDHDFQKKFLALIMFGAFLPIFVSYGIATHNTISDLHHLAKAGMFANMTYINIGTPLHDAIFGSHGTVSDFVAAHAIAGGLHFTMVPLWRMVFFSKVSPWTTKVGMKAKRDGEFPCLGPAYGGTCSISLVDQFYLAIFFSLQVIAPAWFYLDGCWMGSFVATSSEVYKQAAELFKANPTWFSLHAVSNFTSEVTSATSSLKPLVCSNTTMVTWFKPCWAAHFIWAFTFSMLFQYRGSRDEGAMVLKWAHEQVGLGFAGKVYNRALSLKEGKAIGTFLFFKMTVLCMWCLAMV SEQ ID NO:9
 greater than gi|153320 (L05390) hydroxylase [Streptomyces haistedii]
MNARADRAGDTVHRVPVLVVGGSLVGLSTSVFLGRLGVRHMLVERHAGTSVHPRGRGNNVRTMEVYRAAGVEQGIX. SEQ ID NO:10
In compiling the database for screening, standard file formats used to store xe2x80x9cHigh Throughput Sequencingxe2x80x9d programs and other xe2x80x9cGenomexe2x80x9d programs (e.g., FASTA, GENBANK) are read and manipulated. Then, the portions of data which are relevant for the screening step, e.g., the annotation data, are identified. The resulting fields are inserted into oracle. If the nucleotide sequence is a 5xe2x80x2 end, the sequence is directly translated into a peptide sequence, if it is a 3xe2x80x2 end, the complementary sequence of the given sequence is generated and is used for translation (the translation phase takes into account the three possible reading frames to generate the peptide sequence). The last step involves the prediction of the secondary structure of the peptide, which is based on information theory and was developed by J. Garnier, D. Osguthorpe, and B. Robson. The software uses all possible pair frequencies within a window of 17 amino acid residues. After cross validation on a data base of 267 proteins, the predication has a mean accuracy of 64.4% for a three state prediction (helix, beta strand, and coil). The program produces two outputs, one giving the sequence and the predicted secondary structure, the other giving the probability values for each secondary structure at each amino acid position. The predicted secondary structure is the one of highest probability compatible with a helix segment of at least four residues and a extended segment (beta strand) of at least two residues.
Once the relevant data has been assembled in the database, such database can be screened for the presence of a combination of nucleotides or amino acids indicative of the precursor of the peptide comprising the amidated C-terminal end. The purpose of this step of course is to convert the huge amount of unsorted data to a limited set of putative peptides of pharmaceutical interest (potential hormones or hormone fragments, or endogene receptor ligands and the like). The general description of the process is followed by an application to a search of potential amidated peptides in the Expressed Sequence Tags database.
The general process proceeds as follows:
retrieve automated sequences from a public source, such as the internet (EMBL, GENBANK, SWISS-PROT, PDB, etc.)
Analyze file and import data into ORACLE
Select sequences from a subset (e.g. GENBANK EST) or the entire database
search for all of the sequences exhibiting a specific motif of interest, such as precursors of a peptide comprising an amidated C-terminal end
Check that no STOP codon is present in between the AUG codon indicating the beginning of the reading frame and the sought motif;
Select the sequences of unknown biological function
Verify that, when found, the motif is an Open Reading Frame (Kozak consensus sequence)
Compare the environment of the motif location to the one required (e.g., secondary structures required around a maturation site such as the proximity of alpha
helices or beta-sheets.
Check for similarity of the sequences and other known sequences (DEFINITION field)
use threading techniques to search sequences displaying similar secondary structure if no similar structures are defined in the database
If no similar sequence is found, select the sequence as a synthetic candidate whose function has to be determined
If similar sequences of known function are found, the sequences can be selected as a synthetic candidate hose putative function is the one of the similar sequence.
The following examples are given by way of illustration and should in no way be construed as limiting the subject matter disclosed and claimed.